Neurochemical, behavioral, and genetic evidence implicate serotonergic dysfunction in autism, and specifically in restricted and repetitive behaviors (RRBs). Discovery of elevated platelet serotonin (5HT) in ~25-30% of individuals with autism is one of the seminal findings in neuropsychiatric research. Autism is the most heritable complex neuropsychiatric disorder, and platelet 5HT levels are also extremely heritable. Further, elevated platelet 5HT is associated with recurrence risk, both in autism and in obsessive compulsive disorder (OCD). The relationship between autism and OCD is underscored by correlations between RRBs in probands with autism and parental OC symptoms. Recently described mutations in the serotonin transporter (SERT) gene (SLC6A4) result in a common pattern of elevated 5HT transporter activity and a phenotype of autism with RRBs or OCD. Many studies have sought to test for genetic effects at SLC6A4, but the field has been limited by the lack of robust measures of RRBs in autism and key platelet markers of serotonergic function. Indices of Insistence on Sameness (IS) now allow RRB symptoms to be included in genetic analysis. Deciphering the relationships between autism and 5HT requires critical neurochemical phenotypes for elucidation of the underlying mechanisms. For example, our studies of platelet 5HT levels in inbred and outbred populations point to the integrin (53 gene (ITGB3) as a quantitative trait locus for platelet 5HT. Follow-up studies demonstrate association between ITGB3 and autism directly. Preliminary data further show that SERT and ITGB3 physically interact in platelets, and that absence of ItgbS in mouse brain significantly diminishes SERT activity. Other studies find that decreased platelet 5HT2A correlates between boys with autism and their fathers and parallels decreased binding in recent brain studies. We propose further study of SLC6A4, ITGBS and HTR2A to evaluate common and rare variation contributing to a dysregulated 5HT system, IS, and autism. Given strong evidence supporting 5HT involvement generally and these three components specifically, we hypothesize that other variation within 5HT-related genes is very likely to contribute to autism risk as well. We propose to systematically assess the role of 5HT-related genes in autism by using the critical neurochemical and behavioral measures to provide the phenotype data most likely to index genetic liability related to this system